Add ecs-agent container healthcheck localhost ip override env var

README.md

@@ -172,6 +172,7 @@ additional details on each available environment variable.
 | `ECS_ENABLE_TASK_CPU_MEM_LIMIT` | `true` | Whether to enable task-level cpu and memory limits | `true` | `false` |
 | `ECS_CGROUP_PATH` | `/sys/fs/cgroup` | The root cgroup path that is expected by the ECS agent. This is the path that accessible from the agent mount. | `/sys/fs/cgroup` | Not applicable |
 | `ECS_CGROUP_CPU_PERIOD` | `10ms` | CGroups CPU period for task level limits. This value should be between 8ms to 100ms | `100ms` | Not applicable |
+| `ECS_AGENT_HEALTHCHECK_HOST` | `localhost` | Override for the ecs-agent container's healthcheck localhost ip address| `localhost` | `localhost` |
 | `ECS_ENABLE_CPU_UNBOUNDED_WINDOWS_WORKAROUND` | `true` | When `true`, ECS will allow CPU unbounded(CPU=`0`) tasks to run along with CPU bounded tasks in Windows. | Not applicable | `false` |
 | `ECS_ENABLE_MEMORY_UNBOUNDED_WINDOWS_WORKAROUND` | `true` | When `true`, ECS will ignore the memory reservation parameter (soft limit) to run along with memory bounded tasks in Windows. To run a memory unbounded task, omit the memory hard limit and set any memory reservation, it will be ignored. | Not applicable | `false` |
 | `ECS_TASK_METADATA_RPS_LIMIT` | `100,150` | Comma separated integer values for steady state and burst throttle limits for task metadata endpoint | `40,60` | `40,60` |

agent/app/healthcheck_test.go

@@ -17,6 +17,7 @@ import (
 	"fmt"
 	"net/http"
 	"net/http/httptest"
+	"os"
 	"testing"
 	"time"
 
@@ -39,6 +40,22 @@ func TestHealthcheck_InvalidURL2(t *testing.T) {
 	require.Equal(t, 1, rc)
 }
 
+func TestHealthcheck_EnvvarConfig(t *testing.T) {
+	testEnvVar := "TEST_AGENT_HEALTHCHECK_ENV_VAR"
+	defer os.Unsetenv(testEnvVar)
+	os.Setenv(testEnvVar, "127.0.0.1")
+
+	// testing healthcheck url setup from app/run.go
+	localhost := "localhost"
+	if localhostOverride := os.Getenv(testEnvVar); localhostOverride != "" {
+		localhost = localhostOverride
+	}
+	healthcheckUrl := fmt.Sprintf("http://%s:51678/v1/metadata", localhost)
+	expectedUrl := "http://127.0.0.1:51678/v1/metadata"
+
+	require.Equal(t, healthcheckUrl, expectedUrl)
+}
+
 func TestHealthcheck_Timeout(t *testing.T) {
 	sema := make(chan int)
 	ts := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {

agent/app/run.go

@@ -14,6 +14,8 @@
 package app
 
 import (
+	"fmt"
+	"os"
 	"time"
 
 	"github.com/aws/amazon-ecs-agent/agent/app/args"
@@ -24,6 +26,10 @@ import (
 	log "github.com/cihub/seelog"
 )
 
+const (
+	ECS_AGENT_HEALTHCHECK_HOST_ENV_VAR = "ECS_AGENT_HEALTHCHECK_HOST"
+)
+
 // Run runs the ECS Agent App. It returns an exit code, which is used by
 // main() to set the status code for the program
 func Run(arguments []string) int {
@@ -43,7 +49,14 @@ func Run(arguments []string) int {
 		// timeout of 30s. This is so that we can catch any http timeout and log the
 		// issue within agent logs.
 		// see https://docs.docker.com/engine/reference/builder/#healthcheck
-		return runHealthcheck("http://localhost:51678/v1/metadata", time.Second*25)
+		// if ECS_AGENT_HEALTHCHECK_HOST env var is set, it will override
+		// the healthcheck's localhost ip address inside the ecs-agent container
+		localhost := "localhost"
+		if localhostOverride := os.Getenv(ECS_AGENT_HEALTHCHECK_HOST_ENV_VAR); localhostOverride != "" {
+			localhost = localhostOverride
+		}
+		healthcheckUrl := fmt.Sprintf("http://%s:51678/v1/metadata", localhost)
+		return runHealthcheck(healthcheckUrl, time.Second*25)
 	}
 
 	if *parsedArgs.LogLevel != "" {